Cigarette smoking among young adults (ages 18-25) is an important cancer control problem. Despite increasing trends in smoking behavior, this group has been understudied with respect to anti-smoking campaigns. Although much attention has been directed to anti-smoking media campaigns for adolescents, on balance, these efforts have produced modest results. Some have argued that the theoretical underpinnings of such media-based anti-smoking prevention campaigns need to be refined if their efficacy is to be improved. Toward this end, we propose to conduct experimental research to develop and test theory-based anti-smoking video segments (referred to as PSAs) that are targeted to high and low sensation-seeking young adults. The specific aims are to: (1) evaluate the effects of the sensation value of the content of the PSAs in high and low sensation-seekers, (2) to evaluate the effects of the sensation value of the format of the PSAs in high and low sensations-seekers, and (3) to explore the cognitive, affective, and arousal mechanisms by which the different PSAs exert an impact on behavioral intentions in these two groups. Our hypotheses, experimental design, and measures are guided by theories of Communication and Health Behavior Change. Using a laboratory-based experimental design, PSAs will be developed and targeted to high sensation-seekers (HSS) and low sensation-seekers (LSS). Two aspects of the PSAs to be manipulated are sensation value of the content (the theme of the anti-smoking appeal; e.g., tobacco industry manipulation vs. second-hand smoke) and sensation value of the format (the formal features; e.g., high vs. low intensity of visual and auditory effects). Study participants will be 800 males and females ages 18-25 from the Philadelphia area who have smoked at least 1 whole cigarette in the past 6 months. A factorial design will be used to test hypotheses about the impact of 8 different PSAs: (2 per condition): (1) high sensation value (HSV)-content and format; (2) HSV-content, (LSV)-format; (3) LSV-content, HSV-format; and (4) LSV-content and format. Our analyses will focus on the following mediating mechanisms in the communication and behavior change process: (a) arousal; (b) cognitive processing and affective evaluation; (c) comprehension (free and forced recall of content); (d) changes in smoking-related beliefs, attitudes about smoking, beliefs about the consequences, prototype images and social norms; and (e) behavioral intentions and willingness. Those PSAs found to be most effective in our research will be exported to the Legacy Foundation for dissemination and large-scale evaluation.